On Fri, Mar 11, 2005 at 12:27:47PM +0000, Matthew Wilcox wrote:

> > But I guess it's a bit hard to justify adding more infrastructure to
> > support a single callsite which has a simple alternative.  So if you could
> > please add the separate counter?
> 
> It's pretty *small* infrastructure, and it gives me something to whine at
> people about when they use atomic_read on something that isn't an atomic.

Agreed, but I don't mind adding a separate counter.  However...

> If we are going to get rid of sem_getcount, could we rename the 'count'
> variables, at least on i386 and ppc to make it clear that you're not
> supposed to do this ... maybe to 'count_$ARCH'?

I'm working on a patch to do just that.  It fails when building XFS:

fs/xfs/xfs_inode_item.c: In function `xfs_inode_item_pushbuf':
fs/xfs/xfs_inode_item.c:803: error: structure has no member named `count'
fs/xfs/xfs_inode_item.c:825: error: structure has no member named `count'

fs/xfs/linux-2.6/sema.h:
#define valusema(sp)                    (atomic_read(&(sp)->count))

It seems getting the value of a semaphore is more common than it appears
at first glance.  I don't see how this code could possibly work on
parisc.  Therefore I propose:

1. Adding sem_getcount() everywhere, as in my original patch.
2. Renaming count to count_$ARCH as willy suggested.
3. Anyone who abuses semaphores will now break.  Fix them to use
   sem_getcount().

I'll work on that over the weekend unless anyone has any better ideas.

Jody


> 
> -- 
> "Next the statesmen will invent cheap lies, putting the blame upon 
> the nation that is attacked, and every man will be glad of those
> conscience-soothing falsities, and will diligently study them, and refuse
> to examine any refutations of them; and thus he will by and by convince 
> himself that the war is just, and will thank God for the better sleep 
> he enjoys after this process of grotesque self-deception." -- Mark Twain

-- 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to